The invention is particularly suitable for implementing so-called "despun" antennas which are continuous scanning antennas mounted on a satellite that is itself subject to permanent rotary motion about its own axis, and in which the beam of the antenna is scanned at the same speed of rotation as the satellite but in the opposite direction so as to maintain a constant pointing direction in spite of the rotation of the satellite.
Although such a configuration constitutes one of the advantageous implementations of the invention, the invention is itself is in no way limited thereto and, as described below, the teaching of the invention may be applied to a very wide variety of antennas having one or more beams that are electronically controlled.
Similarly, the antenna is described below essentially in terms of transmission, but all of the teaching can be transposed, mutatis mutandis, to operation in reception merely by applying the principle of reciprocity, with the structure of the circuits and their interconnections remaining the same but with the signals traveling from the antenna array towards the transmit/receive circuits instead of traveling in the opposite direction. Under such circumstances, the amplifier stages which are located in the same positions become low-noise amplifier stages with their inputs being connected to the antenna and their outputs being connected to the transmit/receive circuit. Indeed, both types of amplifier (i.e. power amplifiers for transmission and low-noise amplifiers for reception) may coexist in the same module, providing appropriate duplexing or switching is provided.
When radio power is to be radiated (or received) by electronically scanning one or more beams over a wide angular range with optimum efficiency, it is possible to use passive antennas or else to use active antennas.
In essence, so-called "passive" antennas include a main amplifier followed by a fixed or variable power divider together with phase shifters and/or switches.
For transmission, the main drawbacks are: the need to provide a generator of high power (since there is only one amplifier); the occurrence of significant losses downstream from the generator (since the generator is situated upstream from the remainder of the apparatus); and the need to perform switching at high power level. Conversely, on reception, since the low-noise amplifier is situated at the downstream end of the system, the signal is subjected to large losses prior to being amplified, thereby significantly degrading its signal-to-noise ratio.
Finally, and in all cases, having only one amplifier for transmission and/or reception means that a breakdown in the amplifier completely prevents the system from operating since "degraded" mode operation is not possible, i.e. a single fault can completely interrupt the process of transmission or of reception.
An example of one such passive antenna is shown in FIGS. 1 and 2, comprising a circular array 10 having a large number of radiating elements (thirty-two in this example) that are uniformly distributed around a cylindrical surface, as shown diagrammatically in FIG. 2 which is a plan view of the array 10. Successive elements in the circular array are numbered 1 to 32.
The array 10 is fed from a signal source 20. The signal is amplified by a stage 30 and is applied to a beam-forming and scanning network 40, 50 including firstly a power dividing stage 40 and secondly a series of four-way switches 50. In this example, the power dividing stage 40 includes a four-path power divider 41 whose outputs are applied to the inputs of variable two-path dividers 42. The divider 41 is an equal-amplitude and equal-phase fixed divider, whereas the dividers 42 are variable-amplitude equal-phase dividers.
Each of the outputs from the variable power dividers 42 is connected to a four-way switch 50 that feeds four non-contiguous radiating elements in the circular array, which elements are angularly offset from one another at 90.degree. intervals. The output from each divider 42 is thus applied to one of the radiating elements in a subarray, with each subarray being constituted by the four radiating elements having the numbers indicated in the figure (the first subarray is constituted by elements having numbers 1, 9, 17, and 25, the second subarray by elements having numbers 5, 13, 21, and 29, etc.).
By an appropriate combination of variable phase shifts (dividers 42) and switch positions (switches 50), it is possible to cause the beam to scan circularly in a progressive manner: for example the three middle elements (e.g. the elements 2, 3, and 4) are excited in-phase and each with one-fourth of the power, while the remaining fourth is distributed in a manner that varies progressively from one of the outer elements (in this example the element 1) to the other (the element 5) while remaining in phase, thereby obtaining a progressive scan.
This configuration is not free from drawbacks. The main drawback is the very large loss of power between the signal at the output of the amplifier and the signal that is effectively radiated by the array, with this power loss being due to the large number of components passed through. The power loss is generally in the order of 40%.
Another drawback comes from the fact that since scanning is performed by acting on amplitudes only, the phases with which the radiating elements are excited are far from optimum, thereby degrading the quality of the beam.
Another known configuration, as described for example in an article by Boris Sheleg entitled "A matrix-fed circular array for continuous scanning", published in the Proceedings of the IEEE, Vol. 56, No. 11, November, 1968, at pp. 2016 to 2027, uses a single Butler matrix for a similar application.
As shown diagrammatically in FIG. 3, that configuration includes an assembly between the array 10 and the signal source 20 together with its power amplifier 30, which assembly is constituted, from its upstream end to its downstream end by: an equal-amplitude and equal-phase power divider 40 including as many outputs as there are radiating elements; a phase shifting assembly 60 comprising a fixed phase shifter 61 and a variable phase shifter 62 for each of the outputs of the divider 40; and a Butler matrix 70 whose inputs are connected to the outputs of the phase shifters and whose outputs are connected to the various radiating elements of the array 10. (As is known, a Butler matrix is a passive array, theoretically having zero loss, comprising N inputs and N outputs where N is generally a power of 2; the inputs are isolated from one another and a signal applied to any one of the inputs produces currents on all of the outputs, which currents are equal in amplitude but of phase that varies linearly from one element to the next.)
In the apparatus of FIG. 3, scanning is performed by acting on the phase shifters 62 so as to obtain a linear change of phase on the mode inputs while maintaining mode amplitudes that are constant.
Although such a structure eliminates the difficulties associated with switches, it nevertheless suffers from the other drawbacks of the apparatus of FIG. 1.
The second type of antenna is constituted by so-called "active" antennas in which amplification is no longer concentrated at a single point, but is distributed over a plurality of amplifiers.
More precisely, each radiating element is associated with an amplifier connected in the immediate vicinity of the element. The main drawback is that for an antenna having four (or six) facettes, for example, only one amplifier in four (or six) is in use at any given instant, with all of the power being concentrated in the single amplifier associated with the corresponding element in use. This drawback limits the use of this principle to antennas that are required to have a wide scanning range.
In addition, U.S. Pat. No. 4,901,085 in the name of Spring et al. describes a configuration for a multiple beam antenna feed system comprising a plurality of modules forming hybrid matrix power amplifiers. Each of these modules (which are preferably all identical) includes an input matrix and an output matrix having mirror symmetry with each other and interconnected by a battery of power amplifiers. Each of the modules made in this way is connected between a low-level beam-forming network and the radiating elements.
Such a structure requires twice the number of matrices and is thus relatively complex, bulky, and heavy--all of which characteristics are highly disadvantageous for an antenna on board a satellite.
Secondly, in the configuration described in this patent, the beam-forming network connects certain beam-selection ports to certain input ports of the modules, while no signal is applied to certain other ports thereof. As a result the various amplifiers are not identically loaded, and this gives rise to a loss of efficiency in the system.
Finally, and above all, the system described in this prior art does not enable the beam pointing direction to be varied continuously while conserving uniform loading on the amplifiers, whereas this constitutes the essential characteristics of the present invention, as described below.
One of the objects of the present invention is to provide apparatus for electronically controlling the radiation pattern of an electronically-scanned active antenna having one or more beams and operating over a wide angular range with optimum efficiency.
Essentially, this apparatus includes an array of radiating elements subdivided into a number of groups, each beam typically using one or two elements in each group. Amplification takes place in distributed manner using a plurality of amplifiers, with the number of amplifiers being equal to the number of radiating elements, and the connections between the radiating elements and the amplifiers are provided via respective hybrid couplers, means also being provided to optimize and adjust the phases of the signals prior to amplification (in transmission) or after amplification (in reception) so as to control the distribution of energy between the elements.
By applying suitable shifts, this makes it possible to direct the power in the best possible manner to the elements that correspond to the desired pointing direction(s), and to vary power continuously from one portion of the antenna to another so as to change its radiation pattern.
In addition, compared with an active antenna having an amplifier module associated directly with each radiating element, amplification that is distributed in accordance with the present invention has the advantage that power per module can be reduced essentially in the ratio of the number of elements contributing to a beam divided by the total number of elements.
Two advantages are thus obtained: firstly the unit power of the amplifiers is reduced, thereby increasing reliability; and secondly, in the event of one or two of the amplifiers failing, overall performance is little affected by the failure since at any given instant all of the amplifiers in the apparatus ar contributing equally to forming the beam.
In addition, all of the amplifiers are permanently in receipt of signals of equal amplitudes so it is possible to optimize the efficiency of the amplification function.